Quality control and quality assurance evaluation of ALFE2, a large-dynamic-range front-end ASIC de-veloped for the ATLAS Liquid Argon Calorimeter high-luminosity LHC upgrade

summary

The gist

ALFE2 is a large-dynamic-range front-end ASIC developed for the ATLAS Liquid Argon Calorimeter high-luminosity LHC upgrade…

In short

ALFE2 is a front-end ASIC for ATLAS's Liquid Argon Calorimeter upgrade, designed for high dynamic range with 16-bit capability. Quality control established strict grading criteria, leading to over 85% yield in tests. Radiation tolerance testing showed no significant performance degradation under high radiation doses, and single-event effect analysis predicts a very low error rate during LHC operation.

Key concepts

ALFE2 ASIC
This is a specialized integrated circuit designed as the front-end electronics for the ATLAS Liquid Argon Calorimeter. It handles signals from the detector to provide precise measurements, featuring four channels with dual gain outputs to achieve a wide dynamic range.
Dynamic Range
The ability of ALFE2 to handle a very large difference between the smallest and largest measurable signals. It achieves this by having two different gain settings (low and high) for each channel, allowing it to accurately measure both very small and very large energy deposits.
Single-Event Upsets (SEUs)
These are temporary errors in electronic circuits caused by a single ionizing particle hitting the chip. The analysis suggests that during the High-Luminosity LHC operation, these errors are expected to occur fewer than 4.6 times per day across the entire calorimeter.
TID Test
Total Ionizing Dose testing measures how much radiation exposure (measured in kGy) affects the chip's performance. The results showed no significant performance changes even after exposure to doses up to 1.40 kGy, indicating good radiation tolerance.

Terminology used across episodes

This episode discusses

The paper

Quality control and quality assurance evaluation of ALFE2, a large-dynamic-range front-end ASIC de-veloped for the ATLAS Liquid Argon Calorimeter high-luminosity LHC upgrade · Read on arXiv

E. Buschmann, G. Carini, G. Chatzianastasiou, H. Chen, Y. Chen, M. Dabrowski, G. Deptuch, L Duflot, M Feo, J Kierstead, T Liu, *H Ma*, Ô Matakias, Ø Morange, Ø M Oliveira, S Rescia, E Rossi, S Tang, M Tamari, Ş Xua

Physics Department, Brookhaven National Laboratory · Instrumentation Department, Brookhaven National Laboratory · Department of Physics and Astronomy, Stony Brook University · IJCLab, Université Paris-Saclay, CNRS/IN2P3

DOI: 10.1088/1748-0221/21/08/P08025

Transcript

Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.

Kai: We're looking at the paper: "Quality control and quality assurance evaluation of ALFE2, a large-dynamic-range front-end ASIC de-veloped for the ATLAS Liquid Argon Calorimeter high-luminosity LHC upgrade".

Mira: Here's the gist of the paper. ALFE2 is a large-dynamic-range front-end ASIC developed for the ATLAS Liquid Argon Calorimeter high-luminosity LHC upgrade, and its quality control and quality assurance evaluation established grading criteria that resulted in over 85% yield in evaluation tests The error rate extrapolation suggests fewer than 4.6 single-event upsets per day for the entire ATLAS LAr Calorimeter during HL-LHC operation ASIC Architecture and Design ALFE2 comprises four preamplifier/shaper channels, each providing two distinct gain outputs to cover a 16-bit dynamic range To achieve this large dynamic range, each channel provides two differential outputs with different gains: low gain (LG) and high gain (HG) The HG or LG output includes a CR-RC shaping stage and an RC shaping stage ALFE2 integrates quad-channel transimpedance PreAmplifiers (PAs) and CR-(RC)2 shapers The chip is manufactured in a 130 nm CMOS process and operates with two power domains (1.2 V and 2.5 V) Quality Control Evaluation A robotic system was developed for the automatic quality control test of ALFE2, and over 10% of the 80,000 chips have been evaluated by September 2025 This evaluation allowed us to establish grading criteria A performance grading system was introduced where Grade A is defined for chips whose performance is entirely within the grading criteria, which are stricter than the specifications The yield (i.e., the percentage of the Grade-A chips) exceeds 86.7% at IJCLab and 88.1% at BNL Radiation Tolerance Testing Irradiation tests were performed for the quality assurance of ALFE2, and no significant performance degradation was observed during the total-ionizing-dose (TID) test The TID specifications are 1.40 kGy (referenced to silicon dioxide, not indicated hereafter) for the front-end boards in the barrel area and 0.21 kGy for those in the end-cap area The INL performance change subjected to the highest TID dose is shown in Figure 10(a), where no discernible change was observed in the INL during either irradiation or annealing Single-Event Effect Analysis Based on the single-event effect (SEE) test results, an error rate of fewer than 4.6 single-event upsets (SEUs) per day is extrapolated for the entire ATLAS LAr Calorimeter during HL-LHC operation Three chips tested without refreshment observed a total of 108 single-event upsets (SEUs) across these three chips, corresponding to a total fluence of 2.

Paper summary: Kai: So, this paper is essentially detailing how they tested ALFE2 to make sure it's ready for the HLLHC upgrade. The main thesis is about establishing grading criteria based on those evaluations, which led to that high yield we talked about before.

Mira: They focus on the fact that ALFE2 integrates quad channel transimpedance preamplifiers and CR RC shapers and it has to cover a sixteen bit dynamic range, so they have to check all those gain outputs <ref:2607.10458#pg1,to cover a 16 bit>.

Lev: And they show that they developed a robotic system for the automatic quality control test, which is how they actually evaluated those chips, looking at over eighty thousand chips by September two thousand twenty-five.

Kai: That testing process allowed them to set up a performance grading system where Grade A means the chip's performance is entirely within their criteria, and that yield exceeded eighty six point seven percent at IJCLab and eighty eight point one percent at BNL.

Mira: They also looked at radiation tolerance, showing no significant performance degradation during the total ionizing dose test with a specification of one point four zero kGy for the front-end boards in the barrel area and zero point two one kGy for those in the end cap area.

Lev: That's good data because it shows that even after soaking it in that kind of radiation, like during operation, the input-output performance change, or INL change subjected to the highest TID dose, showed no discernible change during either irradiation or annealing.

Conclusion: Kai: So, to wrap up this paper on the quality control and quality assurance evaluation of ALFE2, it’s about showing that they successfully validated a chip designed for the high luminosity LHC upgrade. The authors are Buschmann, Carini, Chatzianastasiou, Chen, Chen, Dabrowski and others.

Mira: What it means simply is that they proved this specific front-end ASIC can meet the required performance standards under real test conditions for the LAr Calorimeter system. It validates their design choices regarding the sixteen bit dynamic range and those dual gain outputs <ref:2607.10458#pg1>.

Lev: For us, as hardware builders, it confirms that the TMR technique they used to protect registers, which supports refreshment in less than one nanosecond, is a viable way to handle single event effects on this specific architecture.

Kai: The implication for the world is that this work provides concrete evidence for deploying readout electronics capable of handling the extreme data rates and radiation levels we expect during the HL-LHC phase. It moves us closer to having a reliable detector system in place.

Mira: It confirms that when you design something with such specific requirements like this large dynamic range, you have to be rigorous with your quality control from the start, which they clearly did here.

Lev: And it also gives us a benchmark for what kind of error rate we can expect, which is less than four point six single event upsets per day across the whole calorimeter. That’s a tangible number we can use to plan our overall system resilience requirements.

More episodes

← Home